DeNoPro - a de novo proteogeomics pipeline to identify clinically relevent novel variants from RNAseq and Proteomics data.

1. Introduction
2. Installation
3. Dependencies
4. Usage
5. GUI

DeNoPro provides a pipeline for the identification of novel peptides from matched RNAseq and MS/MS proteomics data.

The pipeline consists of de novo transcript assembly (Trinity), generation of a protein sequence database of 6-frame translated transcripts, and a combination of search engines (X! Tandem, MS-GF+, Tide) to query the custom database. Identified novel peptides and protein variants are then filtered by confidence and mapped to gene models using ACTG.

To install DeNoPro as a python module, open a terminal in the directory containing setup.py, and run

python setup.py install

DeNoPro can be made executable by running chmod u+x denopro.

DeNoPro has been tested with Python 3, Python 2 is not supported at this time. R version 4.0.0 or greater is required to run the PGA package.

We recommend using a conda environment to maintain dependencies, and an environment config file using Python 3.9.6 and R 4.0.5 has been provided. To setup the conda environment, run conda env create -f denopro-env.yml and activate with conda activate denopro-env.

• Trinity version 2.8.5 - Used during assemble for de novo assembly of RNA transcripts
• PGA (R>4.0) - Used in customdb for creation of 6-frame translated protein database
• PySimpleGUIQt - Used to run the GUI functionality

• SearchGUI version 3.3.17 - Uses the X! Tandem, MS_GF+ and Tide search engines to search created custom database against mgf spectra files
• PeptideShaker version 1.16.42 - Used to select matching identifications among the three search engines to output a list of confident novel peptides and their corresponding proteins
• ACTG - Used to map identified confident novel peptides to their corresponding genomic locations
• Bamstats - Used to process expression levels of novel peptides

DeNoPro was designed to be modular, to account for large processing times. The modes are

assemble : de novo assembly of transcript sequences using Trinity

searchdb : produces custom peptide database from assembled transcripts which are mapped against proteomics data

identify : maps potential novel peptides from searchdb to a reference tracriptome outputting a list of confident novel peptides

novelorf : finds novel ORFs in identified novel peptides

quantify : evaluates expression levels of identified novel peptides in a sample

The standard workflow is assemble >> searchdb >> identify >> novelorf >> quantify

denovo assembly of transcript sequences using Trinity

denopro assemble [options]

• -c/--config_file: Point to the path of config file to use. Default is ./denopro.conf
• --cpu: Maximum number of threads to be used by Trinity
• --max_mem: Maximum number of RAM (in GB) that can be allocated

• output_dir: Directory to use as pipeline output
• dependency_locations/trinity: Full path to Trinity installation
• directory_locations/fastq_for_trinity: Directory containing FASTQ files

Produces custom peptide database from assembled transcripts which are mapped against proteomics data

denopro searchdb [options] 

• -c/--config_file: Point to the path of config file to use. Default is ./denopro.conf

• output_dir: Directory to use as pipeline output
• dependency_locations/searchgui: Full path to SearchGUI .jar file
• dependency_locations/peptideshaker: Full path to PeptideShaker .jar file
• directory_locations/spectra_files: Directory containing .mgf files for database searching
• dependency_locations/hg19: Full path to reference transciptome (FASTA) of protein coding genes

Maps potential novel peptides from customdb to a reference tracriptome, outputting a list of confident novel peptides

denopro identify [options] 

• -c/--config_file: Point to the path of config file to use. Default is ./denopro.conf

• output_dir: Directory to use as pipeline output
• dependency_locations/actg: Full path to directory containing ACTG.jar and param.xml files

Note: Transcriptome model and reference genome are only needed if a serialization file needs to be constructed. If a serialization file is needed, leave serialization_file blank.

• actg_options/transcriptome_gtf: Path to transcriptome model to be used for mapping
• actg_options/ref_genome: Path to directory containing reference genome (each file name must be the same as chromosome number written in the GTF files)
• actg_options/mapping_method: Mapping method to be used. Options are PV (Mapping [P]rotein database first, then [V]ariant splice graph), PS (Mapping [P]rotein database first, then [S]ix-frame translation), VO (Mapping [V]ariant splice graph [O]nly), SO (Mapping [S]ix-frame translation [O]nly)
• protein_database: If mapping_method is PV or PS, path to directory containing protein database
• serialization_file: Path to serialization file of a variant splice graph

Finds novel ORFs in identified novel peptides

denopro novelorf [options]

• -c/--config_file: Point to the path of config file to use. Default is ./denopro.conf

• output_dir: Directory to use as pipeline output

Evaluates expression levels of identified novel peptides

denopro quantify [options]

• -c/--config_file: Point to the path of config file to use. Default is ./denopro.conf

• output_dir: Directory to use as pipeline output
• quantification_options/bamstats: Full path to bamstats .jar file
• quantification_options/bam_files: Full path to directory containing BAM files to be analysed
• quantification_options/bed_file: Full path to BED file to be used. Will be created with data from previous steps if left blank

DeNoPro offers a graphical interface to run the pipeline and edit configuration files.

The GUI uses the Qt framework through PySimpleGUIQt which can be installed with `conda install PySimpleGUIQt'.